No surface-level takes. Real perspectives from 10+ years building automation systems that survive production.
Stop chasing release bugs; start chasing release confidence. GitOps, when coupled with intelligent quality gates, transforms your CI/CD pipeline from a potential failure point into your most reliable release assurance mechanism.
Most people assume Platform Engineering is solely about making developers' lives easier, optimizing their coding and deployment workflows. But for QA professionals, platform engineering is actually about *elevating quality upstream*, transforming us from downstream gatekeepers into architects of frictionless, reliable developer experiences.
Most teams operate under the dangerous illusion that their staging environment is a reliable pre-production replica, when in reality, it’s a Frankenstein's monster of outdated data, mismatched configurations, and synthetic traffic. True release confidence isn't about *having* a staging environment, but about ensuring it's a statistically valid, dynamically refreshed microcosm of production, reflecting real user behavior and system loads.
Most test pipelines are flying blind — they check pass/fail but ignore what the system was actually doing during the test. Wiring OpenTelemetry distributed traces directly into your quality gates turns testing from a binary verdict into a signal-driven feedback loop. Here's how I do it in practice.
OpenAI just paid $1.1 billion for a feature flag company. That's not a deployment story — it's a quality story. Here's why progressive delivery is quietly absorbing what we used to call QA, and what that means for anyone still measuring success by pass rates.
AI isn't replacing QA engineers — it's replacing the repetitive decisions that slow them down. Here's what intelligent automation actually looks like in 2025.
90% of automation projects fail within 18 months. Not because the tools are bad — because the architecture decisions are wrong from day one.
Everyone's talking about self-healing tests. Few have shipped them to production. Here's what actually works, what doesn't, and what it takes to do it right.
Intelligent QA isn't a platform you buy. It's an architecture you build — incrementally, deliberately, and with clear success criteria at each layer.
The jump from tester to architect isn't a title change — it's a complete rewiring of how you see software, systems, and business risk.
Get posts like this in your inbox
No fluff. Sharp takes on QA, AI, and engineering — once a week.